Measurement of Expectancy, Value, and Cost Motivation in Young Ethnically Diverse Student Populations

• In Event: Understanding and Measuring Achievement Motivation in Diverse Contexts Using Practical Measures: Expectancies, Values, and Costs

Abstract

Objective:
A changing economy have focused attention on Science, Technology, Engineering, and Math education in recent years (National Academy of Sciences, 2006; Obama, 2012). Motivation, a key predictor of academic effort, achievement, persistence, and choice, is viewed as an important lever for increasing STEM participation and persistence. This study aims to better understand the motivation of elementary students in science.

Theoretical Framework:
The Expectancy-Value Theory of achievement motivation (Wigfield, Tonks, & Klauda, 2016) integrates social and cognitive factors in understanding student motivation and has provided a framework for understanding student choice, persistence, and achievement in several academic domains, especially math and reading/English (e.g Eccles, Adler, Terry E., Goff, Kaczala, & Midgley, 1983). Research in this area specific to science is most frequently conducted in secondary schools with students who identify as white and middle class. Thus, there is a lack of science motivation research in elementary schools with low-income non-white students.

Method & Data:
To close this gap, this study investigated the factor structure and validity of a nine-item expectancy, value, and cost science motivation instrument in a sample of 1,521 third through fifth graders (51.2% female; 70% black, 23% Hispanic, and 7.8% other race/ethnicity; 94.7% eligible for free- or reduced- price lunch, 11.8% with individualized education plans; 14.0% English language learners). The data were collected as part of a National Science Foundation Math Science Partnership in a large urban school district in the Mid-Atlantic of the United States. Confirmatory factor analysis was utilized to determine the factor structure of the expectancy-value-cost instrument. Structural equation models were used to assess the predictive validity of the measures by regressing state standardized test scores in science, math, and English language arts on the latent factors.

Results & Significance:
Results indicated that the items used in this study reproduced the three-factor structure observed in other similar studies (e.g. Barron & Hulleman, 2015) and reflects expectancy-value theory (Wigfield et al., 2016). Measurement invariance testing demonstrated that the items behaved similarly across gender, race/ethnicity, English language learner status, individualized education plan status, and grade level. Latent mean differences in expectancy, value, and cost indicated decreases of 0.18 to 0.25 standard deviations between third graders and fifth graders in this sample (see Table 1), which aligns with a general decline of motivation with age summarized by Wigfield et al. (2016). Moving to structural models, we predicted each academic outcome using latent expectancy, value, and cost factors (see Table 2). Expectancy was estimated to be a positive predictor of and cost a negative predictor of academic outcomes. The coefficients on the value latent variable were negative and smaller in magnitude than those for cost or expectancy, not surprising as value is theoretically associated with future decisions more than with academic achievement whereas expectancy is theoretically predictive of achievement (Wigfield et al., 2016). The coefficients in the science test scores model were larger in magnitude than in the other models. These results support the use of this instrument in the study of motivation in similar populations of students.

Authors

• David Earle McKinney, Johns Hopkins University

• Jeffrey Grigg, Johns Hopkins University

• Erik Ruzek, University of Virginia

• Carolyn Parker, American University